Archived posting to the Leica Users Group, 2000/04/23

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: [Leica] Kodak versus Fuji: is that the issue?
From: Erwin Puts <imxputs@knoware.nl>
Date: Sun, 23 Apr 2000 20:37:51 +0100

The topic of which film has the better sharpness impression and/or 
lower graininess, and which measurement is the best merit figure to 
evaluate a film is an interesting one and could fill a book to 
elucidate in considerable detail.
Let us first establish some baselines. A film has a number of 
measurable properties, like RMS values (for the random fluctuations 
of grain patterns), MTF-curves for a relationship between contrast 
loss versus spatial frequency, resolution value for the propensity to 
differentiate between closely spaced adjacent point/line objects and 
so on.
At the other end we have perceived, that is subjective, qualities of 
a film. like the impression of sharpness or graininess.
If all would be well, the measurable properties should correlate 
closely to the perceived qualities and more importantly a shift in 
magnitude in one dimension should be proportional to a shift in 
magnitude in the other dimension.
Alas in our real world it is not that simple. Sharpness impression is 
related not to the fineness of the grain pattern, but to acutance, 
this being a property of an emulsion. Acutance can be explained as a 
rapid drop in contrast between two adjacent areas with different 
densities. A grain structure that diffuses the incoming light energy 
has by definition a higher propensity to scatter the light rays and 
so to lower the edge contrast. (= acutance). E-6 emulsions consist of 
dye clouds after development and while these clouds show a very low 
RMS value (low graininess), they also diffuse the light rays and 
lower the acutance. K-films are grain based and while the graininess 
impression is a bit higher they exhibit better edge effects and so 
enhance the sharpness impression. The perception of grain is lower 
when the very fine details are resolved quite good as in this case 
the randomness of the density patterns of the grain clumps or dye 
clouds reside behind the image structure that is recorded.
Looking at MTF graphs we may note that the sharpness impression is 
enhanced if the contrast figures are very high in the 5 to 10 lp/mm 
range. If a film has a value higher than 100% (which should be 
impossible) than the edge contrast is artificially enhanced. But the 
recording capacity of a film is closely related to the contrast value 
at 40 to 50 lp/mm. BUT: differences in contrast value in the 
bandwidth of ±10 percentage points are not relevant. So a film that 
has an MTF value of 40 at% at 50 lp/mm is as good as one with a value 
of 45 to 50%. At the other end of the scale minor differences are 
important and a contrast transfer of 115% is significantly better 
than one of 110% at the same spatial frequency.
When I did a series of tests with K25/64 and Fuji Velvia and 
Provia100F (disregarding the saturation issue here), I found that the 
graininess impression of all four films (when comparing homogeneous 
areas of equal density) was very low and any difference would be 
irrelevant for most photographic purposes with high quality lenses 
(Leica ) and 35mm film at large scale projection. At this level 
differences in graininess are most likely explained by slight 
differences in exposure and with all films every possible detail 
could be recorded.
The K-films however recorded the same details with a better edge 
contrast and details were crisper and  delineated  with  engraving 
like edges. Provia 100F had a definitely softer look, but in all 
fairness had somewhat less detail definition, compared to Velvia and 
the K 25/64.
We are however at a level of definition of details and a quality of 
recording capacity that is at the edge of what a lens can handle and 
the technique of the photographer is presumably the limiting factor 
here.
The overall impression then is that all four films handle the 
recording of object details very well, grain of all four is beyond 
the level of perception, but the K-films record details with more 
edge contrast and   in doing so have a slight advantage. Velvia has 
the finest grain of all four, but as said above the difference with 
K25 and 64 and Provia F is so small as to be irrelevant, but to be 
fair again can be measured. As with speed of a car, you can measure a 
difference between 110 and 112 km/hour, but you can not experience it.
For ultimate recording capacity the K-films still have the edge, it 
is not a very big  one, but in many instances a significant one and I 
would propose that we should not try to establish an absolute 
difference between these four but develop our technique to exploit 
the characteristics of our film/lens combination to suit our goals.
My films of choice are K25 and 64, because of their higher fidelity 
when recording the reality in front of my lens. Comparing K64 with 
P100F at a large screen is no contest: Kodak wins on all visual and 
perceptual counts. In this sense the Velvia and Provia are not in the 
same league, but these films have really awesome capabilities and we 
should not indulge ourselves in an either/or discussion, but 
appreciate the differences as they are and use then when appropriate.
It is really a pity that Kodak has lost all interest in these 
K-series as they could be the cutting edge in emulsion technology. 
But Kodak also has lost interest in the traditional B&W films.

Erwin Puts